Multiscale Similarity Matching for Subimage Queries of Arbitrary Size

Abstract

Many image database management systems support whole-image queries. However, in some situations, users may only remember certain portions of the images. In this paper, we develop Padding and Reduction Algorithms to support subimage queries of arbitrary size based on local color information. The idea is to estimate the best-case lower bound to the distance between the query and the image. To improve the efficiency and effectiveness of content-based retrieval, a multiscale representation is proposed. Since image contents are usually pre-extracted and stored, the number of levels used in such a representation needs to be determined. We address this issue analytically by estimating the CPU and I/O costs, and experimentally by comparing the performance and accuracy of the outcomes of various filtering schemes. Our findings suggest that a 3-level hierarchy is preferred.

We also study three strategies for searching multiple scales. Our studies indicate that the hybrid strategy with horizontal filtering on the coarse level and vertical filtering on remaining levels is the best choice. When used with Padding and Reduction Algorithms in the preferred 3-level multiscale representation, desired images can be retrieved efficiently and effectively.

Keywords

Similarity matching subimage querying multiscale representation cost model search strategy 

References

  1. Bach, J.R. et al. (1996) The Virage Image Search Engine: An Open Framework for Image Management. Proceedings of SPIE Conference on Storage and Retrieval for Still Image and Video Databases IV (Vol. 2670): 76–87. San Jose CA, USA.Google Scholar
  2. Barber, R. et al. (1994) Ultimedia Manager: Query By Image Content and its Applications. Digest of Papers of the Spring COMPCON ‘94: 424–429. San Francisco CA, USA.Google Scholar
  3. Castelli, V. et al. (1997) Searching Image Databases at Multiple Levels of Abstraction. Research Report RC 20702, IBM T. J. Watson Research Center, Yorktown Heights NY, USA.Google Scholar
  4. Chen, J.-Y. et al. (1997) Multiscale Branch and Bound Image Database Search. Proceedings of SPIE Conference on Storage and Retrieval for Image and Video Databases V (Vol. 3022): 133–144. San Jose CA, USA.Google Scholar
  5. Faloutsos, C. et al. (1994) Efficient and Effective Querying by Image Content. Journal of Intelligent Information Systems 3 (3–4): 231–262.CrossRefGoogle Scholar
  6. Faulus, D.S. and Ng, R.T. (1997) An Expressive Language and Interface for Image Querying. Machine Vision and Applications 10 (2): 74–85.CrossRefGoogle Scholar
  7. Flickner, M. et al. (1995) Query by Image and Video Content: The QBIC System. IEEE Computer 28 (9): 23–31.CrossRefGoogle Scholar
  8. Gudivada, V.N. and Raghavan, V.V. (1995) Content-based Image Retrieval Systems. IEEE Computer 28 (9): 18–22.CrossRefGoogle Scholar
  9. Guttman, A. (1984) R-trees: A Dynamic Index Structure for Spatial Searching. Proceedings of ACM SIGMOD Conference on Management of Data: 47–57. Boston MA, USA.Google Scholar
  10. Jacobs, C.E. et al. (1995) Fast Multiresolution Image Querying. Proceedings of ACM SIGGRAPH Conference on Computer Graphics zhaohuan Interactive Techniques: 277–286. Los Angeles CA, USA.Google Scholar
  11. Jain, R., editor (1993) NSF Workshop on Visual Information Management Systems. SIGMOD Record 22 (3): 57–75.Google Scholar
  12. Leung, K.S. (1997) Efficient and Effective Subimage Similarity Matching for Large Image Databases. Master’s Thesis, The University of British Columbia, Vancouver BC, Canada.Google Scholar
  13. Lin, K.-I. et al. (1994) The TV-tree — An Index Structure for High-dimensional Data. VLDB Journal 3 (4): 517–549.CrossRefGoogle Scholar
  14. Ng, R.T. and Tain, D. (1997) An Analysis of Multi-level Color Histograms. Proceedings of SPIE Conference on Storage and Retrieval for Image and Video Databases V (Vol. 3022): 22–34. San Jose CA, USA.Google Scholar
  15. Petkovic, D. et al. (1996) Recent Applications of IBM’s Query By Image Content (QBIC). Research Report RJ 10006, IBM Almaden Research Center, San Jose CA, USA.Google Scholar
  16. Samet, H. (1990) The Design and Analysis of Spatial Data Structures. Addison-Wesley.Google Scholar
  17. Sawhney, H.S. and Hafner, J.L. (1993) Efficient Color Histogram Indexing for Quadratic Form Distance Functions. Research Report RJ 9572, IBM Almaden Research Center, San Jose CA, USA.Google Scholar
  18. Stricker, M. and Dimai, A. (1996) Color Indexing with Weak Spatial Constraints. Proceedings of SPIE Conference on Storage and Retrieval for Still Image and Video Databases IV (Vol. 2670): 29–40. San Jose CA, USA.Google Scholar
  19. Swain, M.J. and Ballard, D.H. (1991) Color Indexing. International Journal of Computer Vision 7 (1): 11–32.CrossRefGoogle Scholar
  20. Thomasian, A. et al. (1997) RCSVD: Recursive Clustering with Singular Value Decomposition for Dimension Reduction in Content-based Retrieval of Large Image/Video Databases. Research Report RC 20704, IBM T. J. Watson Research Center, Yorktown Heights NY, USA.Google Scholar

Copyright information

© Springer Science+Business Media New York 1998

Authors and Affiliations

  1. 1.Department of Computer ScienceThe University of British ColumbiaVancouverCanada

Personalised recommendations